research-from-vault
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions suggest using shell commands like
pdftotextto extract content from PDFs. This introduces a risk of command injection if the agent processes files with adversarial names without proper shell escaping. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the knowledge vault (notes and PDFs) as its primary research source. Malicious instructions hidden in these files could influence agent behavior during synthesis.
- Ingestion points: Markdown files and PDF companions located in the user's configured vault path.
- Boundary markers: Absent. The instructions do not define delimiters or provide warnings for the agent to ignore instructions found within the ingested content.
- Capability inventory: The skill utilizes file system read/write operations and shell command execution (suggested for PDF extraction).
- Sanitization: None specified. The skill relies on standard agent parsing of markdown and text content.
Audit Metadata